Conference Papers

Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506

Browse

Author: search.filters.author.Chittaragi, N.B.Subject: search.filters.subject.Spectral features

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Spectral Feature Based Kannada Dialect Classification from Stop Consonants
    (Springer, 2019) Chittaragi, N.B.; Hegde, P.; Mothukuri, S.K.P.; Koolagudi, G.K.
    This study focuses on the investigation of the significance of stop consonants in view of the classification of Kannada dialects. Majority of the studies proposed have shown the existence of evidential differences in the pronunciation of vowels across dialects. However, consonant based studies on dialect processing are found to be comparatively lesser. In this work, eight stop consonants are used for characterization of five Kannada dialects. Acoustic characteristics such as cepstral coefficients, formant frequencies, spectral flux, and rolloff features are explored from spectral analysis of stops. The consonant dataset is derived from standard Kannada dialect dataset consisting of 2417 consonants obtained from 16 native speakers from each dialect. Support vector machine (SVM) and decision tree-based extreme gradient boosting (XGB) ensemble classification methods are employed for automatic recognition of Kannada dialects. The research findings show that the stops existing for shorter duration also convey dialectal linguistic cues. Combination of spectral properties has contributed to the identification of distinct dialect-specific information across Kannada dialects. © 2019, Springer Nature Switzerland AG.
  • No Thumbnail Available
    Item
    Sentence-Based Dialect Identification System Using Extreme Gradient Boosting Algorithm
    (Springer, 2020) Chittaragi, N.B.; Koolagudi, S.G.
    In this paper, a dialect identification system (DIS) is proposed by exploring the dialect specific prosodic features and cepstral coefficients from sentence-level utterances. Commonly, people belonging to a specific region follow a unique speaking style among them known as dialects. Sentence speech units are chosen for dialect identification since it is observed that a unique intonation and energy patterns are followed in sentences. Sentences are derived from a standard Intonational Variations in English (IViE) speech dataset. In this paper, pitch and energy contour are used to derive intonation and energy features respectively by using Legendre polynomial fit function along with five statistical features. Further, Mel frequency cepstral coefficients (MFCCs) are added to capture dialect specific spectral information. Extreme Gradient Boosting (XGB) ensemble method is employed for evaluation of the system under individual and combinations of features. Obtained results have indicated the influences of both prosodic and spectral features in recognition of dialects, also combined feature vectors have shown a better DIS performance of about 89.6%. © 2020, Springer Nature Singapore Pte Ltd.